We plan to test the hypothesis that immunization regimens optimized for the induction of potent, broadly cross-reactive neutralizing antibodies and cytotoxic T cells (CTL) can confer protective immunity against multiple clades of HIV-1. The program is designed to evaluate the potential efficacy of immunogens that are designed to cross-reactive with predominant strains of HIV-1 prevalent strains of HIV-1 prevalent in Chian, a site of current emergence of a major epidemic of HIV-1 infection. The strategy. Will allow induction of broadly cross-reactive neutralizing antibodies will be developed in Projects 1 and 2. The strategy will involve definition of HIV-1 envelope phenotypes associated with the capacity for induction of such neutralizing antibodies in established small animal models. Immunization will be accomplished by expression of envelopes in vivo using the highly immunogenic, Venezuelan equine encephalitis virus (VEE) replicon system. In vivo env expression will be optimized so as to achieve maximum potency neutralizing antibody responses. Envelopes demonstrated to exhibit the desired immunogenicity phenotypes, or immunotypes, will be compared for immunogenicity when administered using the VEE replicon system, as high quality, soluble oligomeric protein preparations, or in combination. VEE replicon-oligomer prime-boost regimens. Initial studiers will focus on defining clade B envelope characteristics associated with the desired immunotype. In parallel, efforts will proceed to recover envelopes with similar immunotypes from Chinese donors infected with clade B', C', and E strains of HIV-1. Principles established for optimum immunization will be applied to Chinese envelopes. The strategy for induction of potent, cross-reactive CTL will be established in Project 3. Model studies will be conducted using gag, then be extended to tat and to Chinese strains. The studies with the model gag immunogen will involve development of an optimized gene construct, and an optimized gene expression immunization regimen. A number of approaches to gene modification will be pursued, including novel approaches developed by the project PI. Immunogenicity of the optimized gene construct will be determined initially in mice, comparing gene expressed in vivo by plasmid DNA, MVA, and VEE replicons, as well as using prime-boost regimens with plasmid DNA for priming and MVA or VEE for boosting. The nature of T cell epitopes functional in non-human primates in the Chinese gag and tat gene products will be defined. A progressive immunization regimen will be evaluated in rhesus monkeys, determining the cross-reactivity of the neutralizing and CTL responses induced, and efficacy against challenge with pathogenic SHIV. The program is designed to test hypotheses related to development of an immunization regimen that will be suitable for efficacy trials in a complex epidemiologic setting, as is currently present in China.